The Spatial-Numerical Association of Response Codes effect (a.k.a., the SNARC effect) is characterized by the case in which the responses are much faster made by the left hand for relatively small, in magnitude, numbers versus that made by the right hand, whereas the opposite pattern holds for relatively large numbers. To explain the SNARC effect, Dehaene et al. initially proposed the mental number line (MNL) theory, which posited that abstract numbers are represented in the brain via a manner of small numbers on the left while large numbers on the right, forming a left-to-right mental number line. The SNARC effect is the production of the consistent representation between the number’s position on the mental number line and the spatial position response. Series of studies confirmed the key role of spatial references, especially egocentric references of the SNARC effect, they may be flexible whose activation depend on the experimental conditions. However, the spatial coding based on the theory of MNL was limited to the visuospatial information, and it was not able to distinguish the activation conditions of the SNARC effect egocentric references. In addition to the theory of MNL, Gevers et al. (2010) offered a visuospatial and verbal-spatial double coding theory to explain the SNARC effect. In the circumstance of the visuospatial encoding of the theory of MNL, the spatial references of the SNARC effect either activated a manner of body-centered, hand-based, or finger-based. Nevertheless, in the case where both the visuospatial information and the verbal-spatial information work, which pattern of the spatial reference of the SNARC effect will be activated requires further exploration. Therefore, the current study intended to set up three response types via the left hand, right hand, or both hands. We assumed that in the case where both the visuospatial information and the verbal-spatial information work a fashion of body-centered spatial reference will be activated. As expected, neither the intensity of the SNARC effect nor the corresponding number-space mapping patterns showed difference among these response conditions, such a pattern held under the condition that the visuospatial information and verbal-spatial information worked together. These results confirmed that, when both visuospatial information and verbal-spatial information are concerned, a pattern of body-centered spatial reference was activated in the SNARC effect. These findings suggest that the involvement of verbal-spatial information helps to activate the body-centered spatial reference system in the SNARC effect.